The enzymes of the mandelate pathway convert R-mandelate to benzoate which can be further metabolized by the beta-ketoadipate pathway. In preliminary studies, the gene for man- delete racemase from Pseudomonas putida, the first enzyme in the pathway which intercon-verts R- and S-mandelates, has been cloned, sequenced, and expressed to high levels in both Escherichia coli and P. aeruginosa. Using mandelate racemase obtained from the cloned and expressed gene we plan to further characterize the inactivation of the enzyme by a-phenylglycidate, an active site directed irreversible inhibitor. Peptides that are alkylated by the inhibitor will be subjected to automated sequence analysis; these peptides can be placed in the primary sequence deduced from the DNA sequence to reveal the position of presumed active site residues. Random in vitro mutagenesis techniques will be used to generate mutants of the racemase which are resistant to inactivation by alpha-phenylglycidate to determine whether the racemization reaction and inactivation by alpha-aphenylglyidate involve the same functional group. We also plan to clone and map the entire mandelate pathway from P. putida; these studies will provide the gene for benzoyl-formate decarboxylase, the thirdenzyme in the pathway that catalyzes the conversion of benzoylformated to benzaldehyde and CO2; the gene will be sequenced and expressed to high levels. The reactions catalyzed by both mandelate racemase and benzoylformate decarbody-lase involve the intermediacy of carbanionic intermediates; the importance of these intermediates has beren previously established by the use of substrates substituted with p-halomethyl groups. The genes for mandelate racemase and benzoylformate decarboxylase will be subjected to random in vitro mutagenesis to generate mutants that alter the processing of these halosubstituted substrates. These studies will both demonstrate the feasibility of using random mutagenesis techniques to obtain mutants with mechanistically interesting phenotypes and permit the isolation and mechanistic characterization of mutant enzymes in which the lifetime of the carbanionic intermediate has been altered. Finally, the availability of the cloned, sequenced, and expressed genes for mandelated racemase and benzoylformate decarboxylase will permit isolation of large amounts of both proteins for x-ray crystallographic studies in Professor Petsko's laboratory at MIT.